Music generation method and related device

Abstract

A music generation method and a related device, applied to the technical field of computers. In the present application, music that matches a scene in which a vehicle is located can be generated on the basis of input information and an artificial intelligence content generation model, thereby enhancing the richness, creativity, and uniqueness of the generated music, providing with more accurate and convenient experience for users, and satisfying users' requirements for vehicle intelligence and entertainment. Moreover, during music generation, using first speed data of the vehicle or first acceleration data of the vehicle can improve the accuracy of first scene description text, such that the first scene description text more closely matches a first scene in which the vehicle is located, thereby facilitating subsequent generation of music that more closely matches the first scene in which the vehicle is located, and enhancing user immersion and user experience.

Description

A music generation method and related apparatus

[0001] This application claims priority to Chinese Patent Application No. 202411216099.9, filed on August 30, 2024, entitled "A Method for Generating Music and Related Apparatus", the entire contents of which are incorporated herein by reference. Technical Field

[0002] This application relates to the field of computer technology, and in particular to a method and apparatus for generating music. Background Technology

[0003] With the development of intelligent vehicle technology, vehicles are gradually becoming mobile devices integrating driving, intelligence, and entertainment functions. People's demands for vehicle functions in terms of safety, intelligence, entertainment, and convenience are constantly increasing, driving the continuous upgrading of vehicle intelligent cockpit functions. Regarding cockpit entertainment, more and more passengers are playing music while driving. Studies have shown that listening to music while driving can regulate emotions and improve mood; moreover, listening to music can keep the brain active and help improve concentration.

[0004] Currently, music playback in vehicles primarily relies on users selecting their preferred music through the in-vehicle infotainment system, or the system recommending pre-installed music. However, user-selected music can be distracting, affecting vehicle safety, and the frequent selections result in a poor user experience. Music recommendations from the system are typically fixed or preset, lacking variety, creativity, and uniqueness, and the interaction between the user and the vehicle is limited, failing to meet users' needs for vehicle intelligence.

[0005] Therefore, how to play music while users are driving to meet their needs for vehicle intelligence and entertainment is a problem that urgently needs to be solved. Summary of the Invention

[0006] This application provides a music generation method and related apparatus. Based on input information and an artificial intelligence content generation model, this application can generate music that matches the scene in which the vehicle is located, thereby improving the richness, creativity and uniqueness of the generated music, bringing users a more accurate and convenient experience, and meeting people's needs for the intelligence and entertainment of vehicles.

[0007] In a first aspect, this application provides a music generation method, which includes acquiring first input information, obtaining first scene description text based on the first input information, and obtaining first music based on the first scene description text and an artificial intelligence content generation model. The first input information includes first visual data and at least one of the following: first speed data and first acceleration data of a vehicle; the first scene description text indicates the first scene in which the vehicle is located; and the first music includes instrumental music or a song.

[0008] In this application, music that matches the vehicle's current scene is generated based on input information and an AI content generation model. This enhances the richness, creativity, and uniqueness of the generated music, providing users with a more accurate and convenient experience and meeting their needs for vehicle intelligence and entertainment. Furthermore, using the vehicle's first velocity or acceleration data during music generation improves the accuracy of the resulting scene description text, making it more consistent with the vehicle's current environment. This facilitates the generation of music that better matches the scene, enhancing user immersion and overall user experience.

[0009] In addition, music generated through AI content generation models is free from copyright and duplication issues, which can reduce the cost of music generation.

[0010] In one possible implementation of the first aspect, the song includes music and lyrics.

[0011] In the above embodiments, the generated music can also be a song containing lyrics, which can enhance the user's immersion and improve the user experience.

[0012] In another possible implementation of the first aspect, after obtaining the first music based on the first scene description text and the artificial intelligence content generation model, the method further includes acquiring second input information, obtaining a second scene description text based on the second input information, and obtaining the second music based on the second scene description text and the artificial intelligence content generation model. The second input information includes second visual data and at least one of the following: second speed data and second acceleration data of the vehicle; the second scene description text indicates the second scene in which the vehicle is located; the second music includes instrumental music or a song; the first scene is different from the second scene; and the first music is different from the second music.

[0013] Since the vehicle may be in motion, and the surrounding environment may change constantly, the input information needs to be re-acquired and the music regenerated when the vehicle's environment changes. In this way, the music generated by this application can dynamically change with the external environment of the vehicle, providing users with an immersive experience. It can offer personalized listening content for in-car users, giving them a highly immersive and impactful experience. This satisfies people's needs for the intelligence and entertainment of vehicles, improving cabin performance and the user's riding experience.

[0014] In another possible implementation of the first aspect, the difference between the first music and the second music includes at least one of the following: the musical style of the first music is different from that of the second music; the musical elements of the first music are different from those of the second music; the accompaniment of the first music is different from that of the second music; the rhythm of the first music is different from that of the second music; and the lyrics of the first music are different from those of the second music.

[0015] Since music is formed by the combination of multiple musical elements, changes in the scene may cause the regenerated music to differ from the previously generated music. This difference may be due to one or more of these musical elements being different. The above implementation describes some possible differences in musical elements.

[0016] In yet another possible implementation of the first aspect, the first visual data includes a first image of the environment outside the vehicle and / or a second image of the user inside the vehicle.

[0017] In the above embodiments, the first image can be used to extract visual features of the first scene in which the vehicle is located, which can improve the accuracy of the obtained first scene description text. The second image can be used to extract visual features of the user inside the vehicle, which is beneficial for obtaining the user's emotional features subsequently. This allows the music generated later to better meet the user's emotional needs, providing the user with personalized listening content and improving the user experience.

[0018] In another possible implementation of the first aspect, the first input information further includes the vehicle's first ambient temperature data.

[0019] In the above embodiments, using the vehicle's first ambient temperature data during the music generation process can improve the accuracy of the obtained first scene description text, making the first scene description text more consistent with the first scene in which the vehicle is located. This is beneficial for generating music that is more consistent with the first scene in which the vehicle is located, thereby enhancing the user's immersion and improving the user experience.

[0020] In another possible implementation of the first aspect, the first input information further includes first sound data, which includes first sounds of the environment outside the vehicle and / or second sounds of the user inside the vehicle.

[0021] In the above embodiments, the first sound can be used to extract the sound features of the first scene in which the vehicle is located, which can improve the accuracy of the obtained first scene description text. The second sound can be used to extract the sound features of the user inside the vehicle, which is beneficial for obtaining the user's emotional features in the subsequent process. This allows the music generated later to better meet the user's emotional needs, providing the user with personalized listening content and improving the user experience.

[0022] In another possible implementation of the first aspect, the first input information further includes first location data, wherein the musical elements of the first music are related to the musical elements of the region indicated by the first location data.

[0023] Since certain regions may have unique musical elements, in the above implementation, by combining the first location data and considering the musical elements of the region indicated by the first location data during the generation of the first music, users can have an immersive experience, enhance their sense of immersion, and improve their user experience.

[0024] In another possible implementation of the first aspect, obtaining the first scene description text based on the first input information includes obtaining one or more scene features based on the first input information, and obtaining the first scene description text based on one or more scene features and a neural network.

[0025] In the above embodiments, multiple scene features can be combined to obtain the first scene description text, which can improve the accuracy of the first scene description text and make the first scene description text more consistent with the first scene in which the vehicle is located. This is beneficial for the subsequent generation of music that is more consistent with the scene in which the vehicle is located, which can enhance the user's immersion and improve the user experience.

[0026] In another possible implementation of the first aspect, a first scene description text is obtained based on one or more scene features and a neural network, including obtaining the first scene description text based on the scene feature vector of the vehicle at the current moment, the scene feature vector of the vehicle at the previous moment, and the neural network. The scene feature vector of the vehicle at the current moment corresponds to the scene feature vector of the vehicle at the previous moment, and the scene feature vector is related to one or more scene features.

[0027] Since the vehicle may be in motion and the scene in which the vehicle is located may be changing at any time, this application can improve the accuracy of the obtained first scene description text by comparing and analyzing the scene feature vector at the current moment with the scene feature vector of the vehicle at the previous moment.

[0028] In another possible implementation of the first aspect, the time interval between the sampling time of the first input information and the time of the last acquisition of the input information is related to the first speed data of the vehicle, or the sampling duration of the first input information is related to the first speed data of the vehicle.

[0029] For example, as vehicle speed data increases, the interval between two sampling times can be reduced. Similarly, as vehicle speed data increases, the sampling duration for each sample can be increased. Thus, this application adjusts the interval between two sampling times, or the sampling duration for each sample, based on vehicle speed or acceleration data, making sampling more flexible and improving sampling efficiency.

[0030] In another possible implementation of the first aspect, where the first input information includes at least one of a second image of a user inside the vehicle and a second voice of a user inside the vehicle, the method further includes obtaining a first emotion description text based on the first input information, the first emotion description text indicating a first emotion of the user inside the vehicle. Obtaining first music based on a first scene description text and an artificial intelligence content generation model includes obtaining the first music based on the first scene description text, the first emotion description text, and the artificial intelligence content generation model.

[0031] In the above implementation, the user's emotions are taken into account when generating the first piece of music, so that the generated music is more in line with the user's emotional needs, which can provide users with personalized listening content, improve the performance of the cabin and the user's riding experience.

[0032] In another possible implementation of the first aspect, after obtaining the first music based on the first scene description text, the first emotion description text, and the artificial intelligence content generation model, the method further includes obtaining third input information, obtaining a second emotion description text based on the third input information, and obtaining the third music based on the first scene description text, the second emotion description text, and the artificial intelligence content generation model. The third input information includes at least one of a third image of a user inside the vehicle and a third voice of a user inside the vehicle; the second emotion description text indicates a second emotion of the user inside the vehicle; the third music includes instrumental music or a song; the first emotion is different from the second emotion; and the first music is different from the third music.

[0033] Since the vehicle may be in motion, the emotions of the occupants may change constantly. When the occupants' emotions change, the input information needs to be retrieved again and the music needs to be regenerated. In this way, the music generated by this application can dynamically change according to the emotional changes of the occupants, providing personalized listening content and giving users a highly immersive experience. This satisfies people's needs for the intelligence and entertainment of vehicles, and improves the performance of the cabin and the user's riding experience.

[0034] In another possible implementation of the first aspect, obtaining a first emotion description text based on the first input information includes obtaining one or more emotion features based on the first input information, and obtaining the first emotion description text based on one or more emotion features and a neural network.

[0035] In the above embodiments, multiple emotional features can be combined to obtain the first emotional description text, which can improve the accuracy of the first emotional description text and make it more in line with the emotional atmosphere inside the vehicle. This is beneficial for generating music that is more in line with the emotional atmosphere inside the vehicle, thereby enhancing the user's immersion and improving the user experience.

[0036] In another possible implementation of the first aspect, a first emotion description text is obtained based on one or more emotion features and a neural network. This includes obtaining the first emotion description text based on the vehicle's emotion feature vector at the current moment, the vehicle's emotion feature vector at the previous moment, and the neural network. The vehicle's emotion feature vector at the current moment corresponds to the vehicle's emotion feature vector at the previous moment, and the emotion feature vector is related to one or more emotion features.

[0037] Since the vehicle may be in motion, the emotions of the users inside the vehicle may change constantly. This application improves the accuracy of the obtained emotion description text by comparing and analyzing the emotion feature vector at the current moment with the emotion feature vector of the vehicle at the previous moment.

[0038] In another possible implementation of the first aspect, the method further includes obtaining user preference information for the vehicle, the preference information indicating the user's music preference type. Obtaining first music based on a first scene description text and an artificial intelligence content generation model includes obtaining the first music based on the preference information, the first scene description text, and the artificial intelligence content generation model.

[0039] In the above implementation, by combining user preference information, the generated first music is more in line with the user's preferences, which can provide personalized listening content for users in the car and improve the performance of the cabin and the user experience.

[0040] In one possible implementation of the first aspect, obtaining the user's preference information for the vehicle includes receiving an input first instruction, the first instruction being used to indicate the user's preference information, or determining the user's preference information based on an interaction log between the user and the vehicle, the interaction log including at least one of the user's driving information and the vehicle's historical music playback records.

[0041] In another possible implementation of the first aspect, the method further includes acquiring first audio. First music is obtained based on first scene description text and an artificial intelligence content generation model, including obtaining the first music based on the first audio, the first scene description text, and the artificial intelligence content generation model.

[0042] In the above embodiments, the first music is generated by combining the first audio, making the generated first music more in line with the scene in which the vehicle is located, giving the user an immersive feeling and enhancing the user's immersion and experience.

[0043] In another possible implementation of the first aspect, after obtaining the first scene description text based on the first input information, the method further includes displaying a first interface, the first interface including content related to the first scene description text and a first input area, the first input area being used to instruct the user of the vehicle to adjust the content related to the first scene description text.

[0044] In the above implementation, users can adjust the first scene description text through the interactive interface, so that the adjusted first scene description text is more in line with the scene in which the vehicle is located. This is conducive to the subsequent generation of music that is more in line with the scene in which the vehicle is located at the current moment, which can enhance the user's immersion, participation and user experience.

[0045] In another possible implementation of the first aspect, after obtaining the first music based on the first scene description text and the artificial intelligence content generation model, the method further includes displaying a second interface, the second interface including content related to the first music and a first control, the first control being used to adjust at least one of the following: the music style of the first music, the music elements of the first music, the background music of the first music, the music rhythm of the first music, and the lyrics of the first music.

[0046] In the above implementation, users can adjust the first music through the interactive interface, so that the adjusted first music is more in line with the user's preferences and intentions, which can enhance the user's immersion, improve the playability of the generated music, and enhance the user's sense of participation and user experience.

[0047] In another possible implementation of the first aspect, after obtaining the first music based on the first scene description text and the artificial intelligence content generation model, the method further includes obtaining the first video or the fourth image based on the first scene description text and the artificial intelligence content generation model, and displaying a third interface, the third interface including the first video or the fourth image.

[0048] For example, using the first video or fourth image as the background video or image of the first music allows users to view related images or videos while accessing or listening to the first music, giving them an immersive experience. This can enhance user immersion, increase the playability of music generation, improve the attractiveness of the display function, enhance the usability of the display function, and improve user participation and user experience.

[0049] Secondly, this application provides a processing apparatus, which includes an acquisition unit for acquiring first input information, the first input information including first visual data and at least one of the following: first speed data of a vehicle and first acceleration data of a vehicle. The processing unit is configured to obtain first scene description text based on the first input information, and to obtain first music based on the first scene description text and an artificial intelligence content generation model. The first scene description text indicates the first scene in which the vehicle is located, and the first music includes instrumental music or a song.

[0050] In one possible implementation of the second aspect, the song includes music and lyrics.

[0051] In another possible implementation of the second aspect, the acquisition unit is further configured to acquire second input information, which includes second visual data and at least one of the following: second speed data of the vehicle and second acceleration data of the vehicle. The processing unit is further configured to obtain second scene description text based on the second input information, and to obtain second music based on the second scene description text and an artificial intelligence content generation model. The second scene description text is used to indicate the second scene in which the vehicle is located, and the second music includes instrumental music or a song; the first scene is different from the second scene, and the first music is different from the second music.

[0052] In another possible implementation of the second aspect, the difference between the first music and the second music includes at least one of the following: the musical style of the first music is different from that of the second music; the musical elements of the first music are different from those of the second music; the accompaniment of the first music is different from that of the second music; the rhythm of the first music is different from that of the second music; and the lyrics of the first music are different from those of the second music.

[0053] In another possible implementation of the second aspect, the first visual data includes a first image of the environment outside the vehicle and / or a second image of the user inside the vehicle.

[0054] In another possible implementation of the second aspect, the first input information also includes the vehicle's first ambient temperature data.

[0055] In another possible implementation of the second aspect, the first input information further includes first sound data, which includes first sounds of the environment outside the vehicle and / or second sounds of the user inside the vehicle.

[0056] In another possible implementation of the second aspect, the first input information further includes first location data, wherein the musical elements of the first music are related to the musical elements of the region indicated by the first location data.

[0057] In another possible implementation of the second aspect, the processing unit is further configured to obtain one or more scene features based on the first input information, and to obtain a first scene description text based on the one or more scene features and the neural network.

[0058] In another possible implementation of the second aspect, the processing unit is further configured to obtain a first scene description text based on the scene feature vector of the vehicle at the current moment, the scene feature vector of the vehicle at the previous moment, and the neural network. The scene feature vector of the vehicle at the current moment corresponds to the scene feature vector of the vehicle at the previous moment, and the scene feature vector is related to one or more scene features.

[0059] In another possible implementation of the second aspect, the time interval between the sampling time of the first input information and the time of the last acquisition of the input information is related to the first speed data of the vehicle, or the sampling duration of the first input information is related to the first speed data of the vehicle.

[0060] In another possible implementation of the second aspect, the processing unit is further configured to obtain a first emotion description text based on the first input information, the first emotion description text being used to indicate the first emotion of the user inside the vehicle, and to obtain first music based on the first scene description text, the first emotion description text, and the artificial intelligence content generation model.

[0061] In another possible implementation of the second aspect, the acquisition unit is further configured to acquire third input information, which includes at least one of a third image of the user inside the vehicle and a third voice of the user inside the vehicle. The processing unit is further configured to obtain a second emotion description text based on the third input information, and to obtain third music based on a first scene description text, the second emotion description text, and an artificial intelligence content generation model. The second emotion description text is used to indicate the second emotion of the user inside the vehicle, and the third music includes instrumental music or a song; the first emotion is different from the second emotion, and the first music is different from the third music.

[0062] In another possible implementation of the second aspect, the processing unit is further configured to obtain one or more emotion features based on the first input information, and to obtain a first emotion description text based on the one or more emotion features and the neural network.

[0063] In another possible implementation of the second aspect, the processing unit is further configured to obtain a first emotion description text based on the vehicle's emotion feature vector at the current moment, the vehicle's emotion feature vector at the previous moment, and the neural network. The emotion feature vector at the current moment corresponds to the vehicle's emotion feature vector at the previous moment, and the emotion feature vector is related to one or more emotion features.

[0064] In another possible implementation of the second aspect, the acquisition unit is further configured to acquire the user's preference information for the vehicle, the preference information indicating the user's music preference type. The processing unit is further configured to obtain the first music based on the preference information, the first scene description text, and the artificial intelligence content generation model.

[0065] In another possible implementation of the second aspect, the acquisition unit is further configured to receive an input first instruction, the first instruction being used to indicate user preference information, or to determine user preference information based on an interaction log between the user and the vehicle, the interaction log including at least one of the user's driving information and the vehicle's historical music playback records.

[0066] In another possible implementation of the second aspect, the acquisition unit is further configured to acquire the first audio. The processing unit is further configured to obtain the first music based on the first audio, the first scene description text, and the artificial intelligence content generation model.

[0067] In another possible implementation of the second aspect, the processing unit is further configured to display a first interface, the first interface including content related to the first scene description text and a first input area, the first input area being used to instruct the user of the vehicle to adjust the content related to the first scene description text.

[0068] In another possible implementation of the second aspect, the processing unit is further configured to display a second interface, the second interface including content related to the first music and a first control, the first control being configured to adjust at least one of the following: the music style of the first music, the music elements of the first music, the background music of the first music, the music rhythm of the first music, and the lyrics of the first music.

[0069] In another possible implementation of the second aspect, the processing unit is further configured to obtain a first video or a fourth image based on the first scene description text and the artificial intelligence content generation model, and display a third interface, the third interface including the first video or the fourth image.

[0070] Thirdly, embodiments of this application provide a computing device including a processor and a memory, wherein the memory stores a program, and the processor executes the program stored in the memory to enable the computing device to implement the method described in any of the first aspects above.

[0071] Fourthly, this application provides a vehicle that includes the aforementioned processing device or computing device, and the vehicle is used to implement the method described in any of the first aspects.

[0072] Fifthly, embodiments of this application provide a computer-readable storage medium for storing a computer program, the computer program including instructions for performing the method described in any of the first aspects above.

[0073] In a sixth aspect, this application provides a computer program product including computer instructions that, when executed by a processing device, computing apparatus, or processor, cause the method described in any of the first aspects to be implemented.

[0074] The solutions provided in the second to sixth aspects above are used to implement or cooperate with the methods provided in the first aspect above, and therefore can achieve the same or corresponding beneficial effects as the methods in the first aspect, which will not be elaborated here. Attached Figure Description

[0075] The accompanying drawings used in the description of the embodiments will be briefly introduced below.

[0076] Figure 1 is a schematic diagram of the architecture of a vehicle provided in an embodiment of this application;

[0077] Figure 2 is a schematic diagram of the architecture of another vehicle provided in an embodiment of this application;

[0078] Figure 3 is a schematic diagram of the architecture of another vehicle provided in an embodiment of this application;

[0079] Figure 4 is a flowchart illustrating a music generation method provided in an embodiment of this application;

[0080] Figure 5 is a schematic diagram of a sampling method provided in an embodiment of this application;

[0081] Figure 6 is a schematic diagram of another sampling method provided in an embodiment of this application;

[0082] Figure 7 is a schematic diagram of another sampling method provided in an embodiment of this application;

[0083] Figure 8 is a schematic diagram of a first interface provided in an embodiment of this application;

[0084] Figure 9 is a schematic diagram of a scene where a vehicle is located according to an embodiment of this application;

[0085] Figure 10 is a schematic diagram of another scenario where a vehicle is located, according to an embodiment of this application.

[0086] Figure 11 is a schematic diagram of a second interface provided in an embodiment of this application;

[0087] Figure 12 is a schematic diagram of a third interface provided in an embodiment of this application;

[0088] Figure 13 is a schematic diagram of a fourth interface provided in an embodiment of this application;

[0089] Figure 14 is a schematic diagram of a processing device provided in an embodiment of this application;

[0090] Figure 15 is a schematic diagram of the structure of a computing device provided in an embodiment of this application. Detailed Implementation

[0091] The following section will introduce the technical terms that may be used in the embodiments of this application.

[0092] 1. Artificial intelligence (AI) is the theory, methods, technology, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to achieve optimal results. In other words, AI is a branch of computer science that attempts to understand the essence of intelligence and produce a new kind of intelligent machine that can react in a way similar to human intelligence. AI studies the design principles and implementation methods of various intelligent machines, enabling them to have perception, reasoning, and decision-making capabilities. Research in the field of AI includes robotics, natural language processing, computer vision, decision-making and reasoning, human-computer interaction, recommendation and search, and fundamental AI theories. Many scenarios in vehicle autonomous driving systems and infotainment systems require the use of AI model reasoning to provide users with a more accurate and convenient experience, meeting people's needs for the intelligence and entertainment aspects of vehicles.

[0093] 2. Generative Artificial Intelligence Generated Content (AIGC), also known as AI-generated content, refers to the technology of generating relevant content with appropriate generalization ability by learning from and recognizing existing data, based on artificial intelligence techniques such as generative adversarial networks and large-scale pre-trained models. The core idea of ​​AIGC is to use artificial intelligence algorithms to generate content with a certain degree of creativity and quality. AIGC can learn the patterns and structure of input data through training models and large amounts of data, and then generate new content that is similar to the training data but has a certain degree of novelty, rather than simply classifying or predicting data. AIGC can generate text, images, or other media in response to prompts, such as chat generative pre-trained transformers (ChatGPT). For example, by inputting keywords, descriptions, or samples, AIGC can generate matching articles, images, audio, etc.

[0094] 3. A convolutional neural network (CNN) is a type of feedforward neural network whose artificial neurons can respond to a subset of surrounding units within their coverage area, making it excellent for large-scale image processing. A CNN consists of one or more convolutional layers and a fully connected layer at the top (corresponding to a classic neural network), as well as associated weights and pooling layers. This structure allows CNNs to utilize the two-dimensional structure of the input data. Compared to other deep learning architectures, CNNs deliver better results in image and speech recognition.

[0095] 4. A fully connected neural network (FCNN) is a fundamental artificial neural network structure, also known as a multilayer perceptron (MLP). In a fully connected neural network, each neuron is connected to all neurons in the previous and next layers, forming a dense connection structure. Fully connected neural networks can learn complex features of input data and perform tasks such as classification and regression.

[0096] 5. A digital filter is a discrete-time system that filters digital signals to obtain the desired response characteristics.

[0097] 6. Long Short-Term Memory (LSTM) networks are a type of recurrent neural network. Due to their unique design, LSTMs are well-suited for processing and predicting important events in time series with very long intervals and delays.

[0098] 7. Neural networks, in the field of computer science, usually refer to artificial neural networks (ANNs), and are often simply referred to as neural networks or neural network-like systems in engineering and academia. A neural network is an algorithmic mathematical model that mimics the behavioral characteristics of animal neural networks to perform distributed parallel information processing. From an information processing perspective, neural networks abstract the human brain's neuronal network, establishing a simple model and assembling different networks according to different connection methods. This type of network, depending on the complexity of the system, achieves its information processing purpose by adjusting the relationships between a large number of internal nodes (or neurons), and possesses self-learning and adaptive capabilities. Each node represents a specific output function, called the activation function. The connection between each two nodes represents a weighted value for the signal passing through that connection, called a weight, which is equivalent to the memory of the artificial neural network. The network's output varies depending on the network's connection methods, weight values, and activation functions.

[0099] Currently, music playback in vehicles primarily relies on users selecting their preferred music through the in-vehicle infotainment system, or the system recommending pre-installed music. However, user-selected music can be distracting, affecting vehicle safety, and the frequent selections result in a poor user experience. Music recommendations from the system are typically fixed or preset, lacking variety, creativity, and uniqueness, and the interaction between the user and the vehicle is limited, failing to meet users' needs for vehicle intelligence.

[0100] In view of this, this application provides a music generation method and related apparatus that, based on input information and an AIGC model, generates music that matches the scene in which the vehicle is located. This can improve the richness, creativity, and uniqueness of the generated music, providing users with a more accurate and convenient experience. Moreover, music generated through the AIGC model avoids copyright issues, thus reducing the cost of music generation.

[0101] The following section provides an exemplary description of the systems and scenarios in which this application may be applied. It should be noted that the system architecture and business scenarios described in this application are for the purpose of more clearly illustrating the technical solutions of this application and do not constitute a limitation on the technical solutions provided in this application. Those skilled in the art will understand that, with the evolution of system architecture and the emergence of new business scenarios, the technical solutions provided in this application are equally applicable to similar technical problems.

[0102] Please refer to Figure 1, which is a schematic diagram of a vehicle architecture provided in an embodiment of this application. The vehicle 100 includes a data acquisition device 101 and a processing device 102. Of course, the vehicle 100 also includes devices supporting vehicle movement and control, such as a driving system and a braking system, which are not described in detail here. It should be understood that the type of vehicle shown here is only an example. In specific implementations, the vehicle 100 can be different types of vehicles such as automobiles, trucks, trains, buses, vans, and electric vehicles. The various devices included in the vehicle 100 are described below by example:

[0103] The data acquisition device 101 has information acquisition capabilities. For example, the data acquisition device 101 is used to acquire input information, which includes at least one of visual data and vehicle speed data and vehicle acceleration data. For instance, the input information includes visual data and vehicle speed data, or visual data and vehicle acceleration data. Furthermore, the input information includes visual data, vehicle speed data, and vehicle acceleration data.

[0104] For example, visual data includes images of the environment outside the vehicle and / or images of the user inside the vehicle. Optionally, the input information also includes at least one of the vehicle's ambient temperature data, sound data, and location data. The sound data includes sounds from the environment outside the vehicle and / or sounds from the user inside the vehicle.

[0105] For example, the acquisition device 101 may include a vision sensor and one of a speed sensor and an acceleration sensor, and may also include one or more of a temperature sensor, a voice acquisition device, and a positioning device.

[0106] Vision sensors include image sensors (often called cameras) used to acquire image data, including images and videos. An image sensor may contain multiple photosensitive units that sense light signals and form images. Examples of photosensitive units include complementary metal-oxide-semiconductor (CMOS), charge-coupled devices (CCD), and Live MOS. Image sensors are typically classified based on the type of photosensitive units they contain. For example, an image sensor with CMOS photosensitive units is called a CMOS image sensor (CIS). A CIS may include one or more of the following: a color sensor, a monosensor, etc. A color sensor may include a red-green-blue (RGB) sensor. Taking cameras as an example, vision sensors may include monocular cameras, binocular cameras, time-of-flight (TOF) cameras, cameras in driver monitoring systems (DMS), or cameras in cockpit monitoring systems (CMS).

[0107] For example, the acquisition device 101 may include one or more cameras. In this application, the vehicle includes a first camera facing outwards and / or a second camera facing the vehicle cabin. As shown in FIG2, the acquisition device 101 includes a first camera 1011 and a second camera 1012. The first camera 1011 is directed towards the outside of the vehicle 100 and is used to acquire images of the environment outside the vehicle 100. The second camera 1012 is directed towards the cabin of the vehicle 100 and is used to acquire images of the user inside the vehicle 100. Of course, the number and location of the first camera 1011 and the second camera 1012 can be set according to actual needs. For example, the first camera 1011 can be installed on the roof of the vehicle, the sunroof, the headlights, or the windshield. Similarly, the second camera 1012 can be installed near the A-pillar on the driver's side, near the A-pillar on the passenger side, on the headrest (or backrest) of the front seat, on the front of the cabin roof, or integrated into the central control screen. Optionally, the first camera and the second camera may be the same camera. The shooting direction of the camera can be any direction. When the shooting direction of the camera is towards the outside of the vehicle 100, the camera is the aforementioned first camera 1011. When the shooting direction of the camera is towards the cabin of the vehicle 100, the camera is the aforementioned second camera 1012.

[0108] Speed ​​sensors are used to collect vehicle speed data, which can be used to determine whether the vehicle is accelerating, decelerating, or moving at a constant speed.

[0109] Accelerometers are used to collect vehicle acceleration data. This data includes both vertical and horizontal acceleration. Vertical acceleration refers to the vehicle's acceleration in the vertical direction and indicates the degree of bumpiness; it may be related to factors such as the smoothness of the road surface. Horizontal acceleration refers to the vehicle's acceleration in the horizontal direction and indicates its motion state. Optionally, the acceleration data can also be calculated from the vehicle's speed data.

[0110] The temperature sensor is used to collect data on the ambient temperature outside the vehicle.

[0111] A voice acquisition device is used to collect sound; for example, the voice acquisition device can be a microphone. Optionally, the voice acquisition device may include one or more microphones. For example, if the voice acquisition device includes two microphones, the first microphone is located outside the vehicle to collect ambient sound. The second microphone is located inside the vehicle's cabin to collect the voice of the user inside the vehicle. Optionally, the first and second microphones may be the same microphone, used to collect both ambient sound and the user's voice.

[0112] Positioning devices are used to collect vehicle positioning data. For example, positioning devices can be Global Positioning System (GPS) or BeiDou Navigation Satellite System.

[0113] Furthermore, the data acquisition device 101 may also include a humidity sensor for acquiring ambient humidity data outside the vehicle. Optionally, the temperature sensor and humidity sensor may be integrated together, for example, as a temperature and humidity sensor, for acquiring ambient temperature and humidity data outside the vehicle.

[0114] The processing device 102 has data acquisition and data processing capabilities. For example, the processing device 102 can acquire input information collected by the acquisition device 101, obtain scene description text based on the input information, and generate a first piece of music based on the scene description text and an artificial intelligence content generation model. The first piece of music includes instrumental music or a song, and the scene description text indicates the current scene of the vehicle. For example, the scene description text could be: "The vehicle is driving on a coastal road, the sun is shining on the beach, birds are flying in the air and chirping, and the waves are crashing on the beach." For example, the processing device 102 is connected to the acquisition device 101, and can receive data sent (or output) by the acquisition device 101 through the connection line between them, and obtain the scene description text based on the data from the acquisition device 101.

[0115] As one possible implementation, the processing device 102 can be a physical device, such as including one or more of the following modules: a central processing unit (CPU), a microprocessor unit (MPU), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), a complex programmable logic device (CPLD), a coprocessor (assisting the CPU in completing corresponding processing and applications), a microcontroller unit (MCU), a mobile data center (MDC), and / or an electronic control unit (ECU), a cockpit domain controller (CDC), a vehicle integrated / integration unit (VIU), etc. Further, the processing device includes at least one processor integrated in the form of a system-on-chip (SOC), commonly referred to by those skilled in the art as an SOC. The SOC may include at least one processor, and when the SOC includes multiple processors, the types of processors may be different. In addition, a processor can also be a hardware circuit designed for artificial intelligence, which can be understood as a type of ASIC, such as a neural network processing unit (NPU), a tensor processing unit (TPU), a deep learning processing unit (DPU), etc.

[0116] Of course, the above description assumes that the processing device 102 is an in-vehicle device. In some solutions, the processing device 102 can be a physical device located outside the vehicle, such as a server, cloud, or host, which communicates and transmits information with the vehicle. As one possible implementation, the processing device 102 can be a software module, such as a virtual machine, software, program code, or container.

[0117] Optionally, the vehicle 100 may also include a display device 103. The display device 103 has interface display capabilities. The display device 103 includes one or more of the following: a physical screen (such as a central control screen), a projection system, a smart device, or a button panel. Projection systems include, for example, light field screens, head-up displays (HUDs), or other projection systems. As shown in Figure 3, a central control screen 1031 can be installed on the vehicle's central console. The central control screen 1031 is used to display the vehicle's driving route, display configuration information of the vehicle's functions, or play audio and video, etc. In addition to the central control screen, other types of display devices 103 are also installed in the vehicle. As shown in Figure 3, passengers in the front passenger seat, rear passengers, etc., can interact with the vehicle through the display device 103. Of course, the display device 103 (including the central control screen 1031) here includes one or more of the following: a physical screen, a projection system, a smart device, or a button panel. Projection systems include, for example, light field screens, head-up displays (HUDs), or other projection systems. This application does not strictly limit the number and location of the display devices 103 in the applicable scenario. Taking the application of this application in a vehicle as an example, the display devices 103 can be located in front of the passenger seat, on the headrest (or backrest) of the front seat, on the armrest of the seat, on the door, or on the top of the cabin.

[0118] For example, the display device 103 is connected to the processing device 102. The display device 103 can receive content related to the first music sent (or output) by the processing device 102 through the connection line between the two, and display the content related to the first music through an interface. For example, the display device 103 receives a second interface sent (or output) by the processing device 102. The second interface includes content related to the first music and a first control. The content related to the first music includes the lyrics of the first music, the playback volume of the first music, the playback sound effects of the first music, etc. The first control is used to adjust at least one of the following: the music style of the first music, the music elements of the first music, the background music of the first music, the rhythm of the first music, and the lyrics of the first music.

[0119] In one possible implementation, the processing device 102 acquires the input information collected by the acquisition device 101, obtains scene description text based on the input information, and generates first music based on the scene description text and an artificial intelligence content generation model. The scene description text indicates the scene in which the vehicle is located. This application can generate music that matches the scene in which the vehicle is located based on input information and an artificial intelligence content generation model, thereby improving the richness, creativity, and uniqueness of the generated music, providing users with a more accurate and convenient experience, and meeting people's needs for the intelligence and entertainment of vehicles.

[0120] The methods of the embodiments of this application will be described in detail below.

[0121] Please refer to Figure 4, which is a schematic flowchart of a music generation method provided in an embodiment of this application. Optionally, this method can be applied to a vehicle, such as the vehicle 100 shown in Figure 1, and can optionally be executed by the processing device 102 in the vehicle 100.

[0122] The music generation method shown in Figure 4 may include steps S401-S403. Steps S401-S403 are detailed below:

[0123] Step S401: The processing device acquires the first input information.

[0124] The processing device is a device with data processing capabilities. For example, the processing device is the processing device 102 shown in FIG1 above.

[0125] The first input information includes first visual data, and further includes at least one of the following: first vehicle speed data and first vehicle acceleration data. For example, the first input information includes first visual data and first vehicle speed data, or the first input information includes first visual data and first vehicle acceleration data. As another example, the first input information includes first visual data, first vehicle speed data, and first vehicle acceleration data.

[0126] For example, the first visual data includes a first image of the environment outside the vehicle and / or a second image of the user inside the vehicle.

[0127] Optionally, the first input information may also include at least one of the following: first ambient temperature data of the vehicle, first sound data, and first location data. The first sound data includes first sounds of the external environment of the vehicle, and / or, second sounds of the user inside the vehicle.

[0128] In one possible implementation, the vehicle includes a data acquisition device for acquiring input information, such as first input information. A processing device is connected to the data acquisition device and can receive the first input information sent (or output) by the data acquisition device through the connection line between the two, thereby acquiring the first input information.

[0129] For example, the acquisition device is the acquisition device 101 shown in Figure 1 above. For example, the acquisition device includes a vision sensor and one of a speed sensor and an acceleration sensor, and may also include one or more of a temperature sensor, a voice acquisition device, and a positioning device. For example, the vision sensor includes a first camera used to acquire images of the environment outside the vehicle; for example, the first camera is the first camera 1011 shown in Figure 2 above. The speed sensor, acceleration sensor, temperature sensor, voice acquisition device, and positioning device are described in the foregoing related descriptions.

[0130] Optionally, the time interval between the sampling time of the first input information and the time of the last acquisition of input information is related to the vehicle's first speed data. Here, sampling refers to the processing device acquiring the input information, or the acquisition device acquiring the input information.

[0131] For example, the first input information is instantaneous sampling information, that is, information sampled at a certain moment. For instance, the first input information includes a first image of the vehicle's external environment, the vehicle's first ambient temperature data, and the vehicle's speed data—information at a specific moment. As shown in Figure 5, when the vehicle's first speed data is K1, the interval between two sampling moments is T1, for example, the interval between the nth sampling moment and the (n+1)th sampling moment is T1. When the vehicle's first speed data is K2, K1 is less than K2, and the interval between two sampling moments is T2, for example, the interval between the mth sampling moment and the (m+1)th sampling moment is T2. That is, as the vehicle's speed increases, the interval between two sampling moments can decrease; that is, as the vehicle's speed increases, the sampling speed also increases. Optionally, the interval between two sampling moments can be calculated based on the sampled information and the vehicle's speed data.

[0132] Alternatively, the sampling duration of the first input information is related to the vehicle's first speed data.

[0133] For example, the first input information is time-sampling information, that is, information sampled within a certain duration, with no interval between two samples, meaning the sampling is continuous. For instance, the first input information includes video data and audio data within a certain duration. As shown in Figure 6, when the vehicle's first speed data is K1, the sampling duration for each sample is t1, with no interval between two samples; for example, the continuous sampling duration for the nth sample is t1. When the vehicle's first speed data is K2, K1 is less than K2, and the sampling duration for each sample is t2, with no interval between two samples; for example, the continuous sampling duration for the mth sample is t2. In other words, as the vehicle's speed increases, the sampling duration for each sample can increase, and the interval between two samples can decrease; that is, as the vehicle's speed increases, the continuous sampling duration also increases, and the sampling speed also increases. Optionally, the continuous sampling duration for each sample can be calculated based on the sampled information and the vehicle's speed data.

[0134] Alternatively, the time interval between the sampling time of the first input information and the time of the last acquisition of input information is related to the vehicle's first speed data, and the sampling duration of the first input information is related to the vehicle's first speed data.

[0135] For example, the first input information is time-sampling information, that is, information sampled within a certain time period, with an interval between two samples. As shown in Figure 7, when the vehicle's first speed data is K1, the sampling duration of each sample is t3, and the interval between two samples is T3. For example, the continuous sampling duration for the nth sample is t3, and the interval between the nth and (n+1)th sampling times is T3. When the vehicle's first speed data is K2, K1 is less than K2, the sampling duration of each sample is t4, and the interval between two samples is T4. For example, the continuous sampling duration for the mth sample is t4, and the interval between the mth and (m+1)th sampling times is T4. That is, as the vehicle's speed data increases, the sampling duration of each sample can increase; that is, as the vehicle's speed increases, the continuous sampling duration also increases. Optionally, the continuous sampling duration of each sample and the interval between two samples can be calculated based on the sampled information and the vehicle's speed data.

[0136] Step S402: The processing device obtains the first scene description text based on the first input information.

[0137] The first scene description text is used to indicate the first scene in which the vehicle is located. For example, the first scene description text is used to indicate the first scene in which the vehicle is stationary, or to indicate the first scene in which the vehicle is located at a certain moment while it is in motion.

[0138] In one possible implementation, the processing device obtains one or more scene features based on the first input information, and obtains scene description text based on one or more scene features and a neural network.

[0139] Since the first input information includes at least two of the multiple pieces of information, the processing device can determine that there are multiple possible scenarios for the scene features based on the first input information. Examples of several possible scenarios are described below.

[0140] Example 1: The first input information includes first visual data and first vehicle speed data.

[0141] For example, the first visual data includes a first image of the environment outside the vehicle. The processing device obtains visual features of the vehicle's exterior based on the first visual data and a convolutional neural network. These visual features belong to one or more of the aforementioned scene features. Since the image data is a two-dimensional data matrix with significant correlation between adjacent elements, using a convolutional neural network for feature extraction can not only improve the accuracy of the obtained visual features but also increase the speed of feature extraction. For instance, the processing device obtains the visual features of the vehicle's exterior as "seaside, beach, sunshine, road, birds, and waves" based on the first visual data and the convolutional neural network.

[0142] For another example, the processing device obtains the vehicle's first motion feature based on the vehicle's first speed data and a long short-term memory neural network. This first motion feature belongs to one of the aforementioned scene features. Since the vehicle's speed data may change constantly, extracting the vehicle's motion feature requires relying on historical speed data. Using a long short-term memory network for feature extraction not only improves the accuracy of the obtained motion features by utilizing historical speed data but also increases the speed of feature extraction. For example, the processing device obtains the vehicle's first motion feature as "the vehicle's speed is 100 km / h, and the vehicle is accelerating" based on the vehicle's first speed data and the long short-term memory neural network.

[0143] Example 2: The first input information includes first visual data and first acceleration data of the vehicle.

[0144] For details on extracting scene features from first-visual data, please refer to the relevant description in Example 1 above.

[0145] For example, the processing device obtains the vehicle's second motion characteristic based on the vehicle's first acceleration data and a long short-term memory neural network. The vehicle's first acceleration data includes the vehicle's first vertical acceleration data and the vehicle's first horizontal acceleration data. The first vertical acceleration data refers to the vehicle's acceleration in the vertical direction, used to indicate the degree of bumpiness of the vehicle, and may be related to information such as the smoothness of the road the vehicle is currently traveling on. The first horizontal acceleration data refers to the vehicle's acceleration in the horizontal direction, used to indicate the vehicle's motion state. Optionally, the vehicle's acceleration data can also be calculated based on the vehicle's speed data. The vehicle's second motion characteristic belongs to one or more of the aforementioned scene features.

[0146] Since vehicle acceleration data can change constantly, extracting vehicle motion features requires relying on historical acceleration data. Using a Long Short-Term Memory (LSTM) network for feature extraction not only improves the accuracy of the obtained motion features by utilizing historical acceleration data but also increases the speed of feature extraction. For example, based on the vehicle's first acceleration data and the LTM neural network, the processing device obtains the vehicle's second motion feature as "the vehicle's vertical acceleration is 0.25 * 9.8 m / s²". 2 "The vehicle is in a state of severe shaking." For example, based on the vehicle's first acceleration data and a long short-term memory neural network, the processing device obtains the vehicle's second motion characteristic as "the vehicle's horizontal acceleration is 6 m / s²". 2 The vehicle is in a state of rapid acceleration.

[0147] Example 3: The first input information includes at least one of the following: first visual data and first speed data and first acceleration data of the vehicle, and also includes first ambient temperature data of the vehicle.

[0148] The extraction of scene features using first visual data, first vehicle speed data, and first vehicle acceleration data can be found in the relevant descriptions in Examples 1 and 2 above.

[0149] For example, the processing device obtains a first environmental characteristic of the vehicle based on the vehicle's first ambient temperature data and a digital filter. This first environmental characteristic belongs to one or more of the aforementioned scene characteristics. For instance, the processing device obtains the vehicle's first environmental characteristic as "ambient temperature 30°C" based on the vehicle's first ambient temperature data and the digital filter.

[0150] Example 4: The first input information includes at least one of the following: first visual data and first speed data and first acceleration data of the vehicle, and also includes first ambient humidity data of the vehicle.

[0151] The extraction of scene features using first visual data, first vehicle speed data, and first vehicle acceleration data can be found in the relevant descriptions in Examples 1 and 2 above.

[0152] For example, the processing device obtains a second environmental feature of the vehicle based on the vehicle's first ambient humidity data and a digital filter. This second environmental feature belongs to one or more of the aforementioned scene features. Since ambient temperature and humidity data change relatively slowly, feature extraction can be achieved by using a digital filter to remove outliers, thus improving the speed of feature extraction. For instance, based on the vehicle's first ambient humidity data and the digital filter, the processing device obtains the vehicle's second environmental feature as "ambient humidity 80%".

[0153] Example 5: The first input information includes at least one of the following: first visual data and first speed data and first acceleration data of the vehicle, and also includes first sound data.

[0154] The extraction of scene features using first visual data, first vehicle speed data, and first vehicle acceleration data can be found in the relevant descriptions in Examples 1 and 2 above.

[0155] For example, the first sound data includes first sound data of the environment outside the vehicle. The processing device obtains the sound features of the environment outside the vehicle based on the first sound of the environment outside the vehicle and a long short-term memory neural network. The sound features of the environment outside the vehicle belong to one or more of the aforementioned scene features. Since sound data has a strong time dependency, using a long short-term memory network for feature extraction can capture this time dependency, handle variable-length input sequences, and remember long-term information dependencies, thereby improving the accuracy of the obtained sound features of the environment outside the vehicle and increasing the speed of feature extraction. For example, the processing device obtains the sound features of the environment outside the vehicle as "ocean waves and birdsong" based on the first sound data and the long short-term memory neural network.

[0156] Example 6: The first input information includes at least one of the following: first visual data and first speed data and first acceleration data of the vehicle, and also includes vibration data of the vehicle.

[0157] The extraction of scene features using first visual data, first vehicle speed data, and first vehicle acceleration data can be found in the relevant descriptions in Examples 1 and 2 above.

[0158] Optionally, the acquisition device also includes a vibration sensor for collecting vehicle vibration data. For example, the processing device can receive the vehicle vibration data collected by the acquisition device and obtain the vehicle's vibration characteristics based on the vehicle vibration data and a fully connected neural network. The vehicle vibration characteristics belong to one or more of the aforementioned scene features. Since vibration data is a one-dimensional continuous signal with no dependency between adjacent signals, using a fully connected neural network for feature extraction can not only improve the accuracy of the obtained vibration features but also increase the speed of feature extraction. Optionally, the vehicle's vibration characteristics can be represented by different vibration levels, with a higher vibration level indicating a more pronounced vibration amplitude. For example, taking vibration level 3 as the largest vibration amplitude and vibration level 1 as the smallest vibration amplitude, the processing device, based on the vehicle vibration data and the fully connected neural network, obtains the vehicle's vibration characteristics as "intense vibration, vibration level 3".

[0159] The above describes several possible examples of how processing devices obtain scene features based on the first input information. It should be understood that there may be more examples in the actual implementation process. In addition, multiple examples can be combined, which will not be listed here.

[0160] Furthermore, the processing device obtains a first scene description text based on one or more extracted scene features and a neural network. By combining multiple scene features to obtain the first scene description text, the processing device can improve its accuracy, making it more consistent with the scene in which the vehicle is located. This facilitates the subsequent generation of music that better matches the scene, enhancing user immersion and improving the user experience.

[0161] For example, the processing device obtains a first scene description text based on the vehicle's scene feature vector at the current moment, the vehicle's scene feature vector at the previous moment, and a neural network. The scene feature vector at the current moment corresponds to the scene feature vector at the previous moment, and the scene feature vector is related to one or more scene features. For example, the scene feature vector at the current moment includes the visual features of the vehicle's exterior and the sound features of the vehicle's external environment at the current moment; similarly, the scene feature vector at the previous moment includes the visual features of the vehicle's exterior and the sound features of the vehicle's external environment at the previous moment. For example, the visual features of the vehicle's exterior might be "seaside, beach, sunshine, road, birds, and waves," and the sound features of the vehicle's external environment might be "wave sounds and birdsong." Based on the visual features of the vehicle's exterior at the current moment, the sound features of the vehicle's external environment at the current moment, the visual features of the vehicle's exterior at the previous moment, the sound features of the vehicle's external environment at the previous moment, and the neural network, the processing device obtains the first scene description text as "The vehicle is driving on the coastal road, the sun shines on the beach, birds fly in the air and make bird calls, and the waves crash on the beach and make the sound of waves".

[0162] Optionally, the scene feature vector of the vehicle at the current moment can be obtained by the processing device by fusing one or more scene features of the vehicle at the current moment with a fully connected neural network. The scene feature vector of the vehicle at the previous moment can also be obtained by the processing device by fusing one or more scene features of the vehicle at the previous moment with a fully connected neural network. Since the vehicle may be in motion, the scene in which the vehicle is located may change constantly. By comparing and analyzing the scene feature vector at the current moment and the scene feature vector at the previous moment, the processing device can improve the accuracy of the obtained scene description text. For example, through the above implementation method, this application can make the obtained first scene description text more consistent with the scene in which the vehicle is located at the current moment, and the obtained scene description text can be continuously updated as the scene in which the vehicle is located changes. This is beneficial for generating music that is more consistent with the scene in which the vehicle is located at the current moment, thereby enhancing the user's immersion and improving the user experience.

[0163] As one possible implementation, if the first input information includes at least one of a second image of a user inside the vehicle and a second voice of a user inside the vehicle, the processing device can obtain a first emotion description text based on the first input information. The first emotion description text is used to indicate the first emotion of the user inside the vehicle.

[0164] Similarly, the processing device obtains one or more emotion features based on the first input information, and obtains a first emotion description text based on one or more emotion features and a neural network.

[0165] For example, the first input information includes a second image of a user inside the vehicle. The processing device obtains visual features of the vehicle interior based on the first input information and a convolutional neural network. These visual features belong to one or more of the aforementioned emotional features. For instance, the processing device obtains the visual features of the vehicle interior as "the user's face is beaming with joy and their posture is laughing heartily" based on the first input information and the convolutional neural network.

[0166] For another example, the first input information includes the second voice of a user inside the vehicle. The processing device obtains the voice characteristics of the user inside the vehicle based on the first input information and a long short-term memory neural network. The voice characteristics of the user inside the vehicle belong to one or more of the aforementioned emotional characteristics. For example, the processing device obtains the voice characteristic of the user inside the vehicle as "loud laughter" based on the first input information and the long short-term memory neural network.

[0167] Furthermore, the processing device obtains a first emotional description text based on one or more extracted emotional features and a neural network. Combining multiple emotional features to obtain the first emotional description text improves its accuracy, making it more consistent with the emotional atmosphere inside the vehicle. This facilitates the subsequent generation of music that better matches the vehicle's emotional atmosphere, enhancing user immersion and improving the user experience.

[0168] For example, the processing device obtains a first emotion description text based on the vehicle's emotion feature vector at the current moment, the vehicle's emotion feature vector at the previous moment, and a neural network. The vehicle's emotion feature vector at the current moment corresponds to the vehicle's emotion feature vector at the previous moment, and the emotion feature vector is related to one or more emotion features. For example, the vehicle's emotion feature vector at the current moment includes the visual features inside the vehicle at the current moment and the voice features of the users inside the vehicle at the current moment; then, the vehicle's emotion feature vector at the previous moment includes the visual features inside the vehicle at the previous moment and the voice features of the users inside the vehicle at the previous moment. For example, the visual features inside the vehicle might be "a broad smile and a hearty laugh," and the voice features of the users inside the vehicle might be "loud laughter." Based on the visual features inside the vehicle at the current moment, the voice features of the user inside the vehicle at the current moment, the visual features inside the vehicle at the previous moment, the voice features of the user inside the vehicle at the previous moment, and the neural network, the processing device obtains the first emotion description text as "the user's face is beaming with joy, his posture is laughing heartily, and he is emitting a loud laugh".

[0169] Optionally, the vehicle's emotion feature vector at the current moment can be obtained by the processing device through the fusion of one or more emotion features of the vehicle at the current moment and a fully connected neural network. The vehicle's emotion feature vector at the previous moment can also be obtained by the processing device through the fusion of one or more emotion features of the vehicle at the previous moment and a fully connected neural network.

[0170] Since the vehicle may be in motion, the emotions of the users inside the vehicle may change constantly. By comparing and analyzing the emotion feature vector at the current moment with the emotion feature vector at the previous moment, the processing device can improve the accuracy of the obtained emotion description text. For example, through the above-described implementation, this application can make the obtained first emotion description text more consistent with the emotions of the users inside the vehicle at the current moment. Moreover, as the emotions of the users inside the vehicle change, the obtained emotion description text can be continuously updated, which is beneficial for generating music that is more consistent with the emotions of the users inside the vehicle at the current moment, thereby enhancing the user's immersion and improving the user experience.

[0171] Furthermore, after obtaining the first scene description text, the processing device can output the obtained first scene description text or content related to the first scene description text to the user through an interactive device. The interactive device is an output device connected to the computing device, which can be used to output the first scene description text or content related to the first scene description text. Optionally, the output modality of the first scene description text or content related to the first scene description text includes at least one of visual output, auditory output, and tactile output, and the interactive device used to output the first scene description text or content related to the first scene description text includes one or more of a screen, a light strip, an audio playback device, and a tactile output device.

[0172] As an example of visual output, an interactive device that outputs the first scene description text or content related to the first scene description text may include a screen that can display an interface, such as the central control screen 1031 shown in Figure 3. The interface can output the first scene description text or content related to the first scene description text in the form of text, symbols, or graphics.

[0173] For example, after obtaining the first scene description text, the processing device can generate a first interface and control the vehicle's interactive device to display the first interface. The first interface includes content related to the first scene description text and a first input area. The first input area is used to instruct the vehicle user to adjust the content related to the first scene description text. As shown in Figure 8, the first interface includes the first scene description text and the first input area. The first scene description text is "The vehicle is driving on a road by the sea, the sun shines on the beach, birds fly in the air and chirp, and the waves crash against the beach." The first input area is blank. The interactive device can input the user's voice information (e.g., "Add a water column from a whale that looks like fireworks") into the first input area, that is, the content in the first input area is updated to "Add a water column from a whale that looks like fireworks." The interactive device adjusts the content of the first scene description text, resulting in the user-adjusted first scene description text: "The vehicle is driving on a coastal road, the sun shines on the beach, birds fly in the air and sing, the waves crash against the shore, and the water spouts from whales are as beautiful as fireworks." In this way, the user can adjust the first scene description text obtained by the processing device through the interactive device, making the adjusted text more consistent with the vehicle's current environment. This facilitates the generation of music that better matches the current scene, enhancing the user's immersion and improving the user experience.

[0174] Optionally, users can input information in the first input area via voice input or touch screen input, thereby adjusting the content related to the first scene description text.

[0175] Furthermore, the processing device acquires the user-adjusted first scene description text so that music that is more consistent with the scene in which the vehicle is located can be generated based on the user-adjusted first scene description text.

[0176] Similarly, after receiving the first emotion description text, the processing device can generate a fifth interface and control the vehicle's interactive device to display it. The fifth interface includes content related to the first emotion description text and a second input area. The second input area instructs the vehicle user to adjust the content related to the first emotion description text. In this way, the user can adjust the first emotion description text obtained by the processing device through the interactive device, making the adjusted text more consistent with the user's current emotion. This facilitates the generation of music that better matches the user's current mood, enhancing immersion and improving the user experience.

[0177] Furthermore, the processing device acquires the user-adjusted first emotion description text so that music that better matches the user's emotions inside the vehicle can be generated based on the user-adjusted first emotion description text.

[0178] Step S403: The processing device generates the first music based on the first scene description text and the artificial intelligence content generation model.

[0179] The first music can be either instrumental music or a song, and a song can include both music and lyrics. In other words, the first music generated in this application can be instrumental music consisting only of music, or it can be a song that includes both lyrics and music.

[0180] Music is usually composed of multiple musical elements. For example, the first piece of music includes musical style (or genre), accompaniment, musical elements, or rhythm. When the first piece of music is a song, it also includes the musical element of lyrics.

[0181] In one possible implementation, the processing device can obtain at least one of the following: the musical style of the first music, the background music of the first music, the musical elements of the first music, the rhythm of the first music, and the lyrics of the first music, based on the first scene description text and the artificial intelligence content generation model, and then obtain the first music. Optionally, the processing device can obtain the musical style of the first music, the musical elements of the first music, the rhythm of the first music, the background music of the first music, and the lyrics of the first music in sequence, and then combine them to obtain the first music.

[0182] For example, the first scene description text is "Vehicles are driving on the seaside highway, the sun shines on the beach, birds fly in the air and chirp, and the waves crash against the beach." Based on this first scene description text and the artificial intelligence content generation model, the processing device determines that the first music's musical style is "pop music," its musical elements are "guitar sounds and wave sounds," its accompaniment is "upbeat melody," its rhythm is brisk, and its lyrics include one or more of the following: "seaside, highway, sunshine, beach, birds, flying, birdsong, waves." Further, based on the aforementioned musical style, musical elements, accompaniment, rhythm, lyrics, and the artificial intelligence content generation model, the processing device obtains the first music.

[0183] Optionally, each of the musical style, musical elements, background music, rhythm, and lyrics of the first music can be generated by a separate AI content generation model or by the same AI content generation model; there are no restrictions here.

[0184] Of course, in the process of generating the first piece of music, the processing device may need to use other data besides the scene description text. The following is an example of other data.

[0185] For example, during the generation of the first music, the processing device can also incorporate the emotional description text of the vehicle's user. For instance, the processing device obtains the first music based on the first scene description text, the first emotional description text, and the artificial intelligence content generation model. The first emotional description text is described in the relevant section of step S402. Optionally, the processing device adjusts the music style or background music bias of the first music based on the vehicle's user's emotional description text. For example, if the vehicle's user's emotional description text indicates that the user's current mood is high, excited, or happy, then the music style or background music of the first music is adjusted to a bright bias. Similarly, if the vehicle's user's emotional description text indicates that the user's current mood is low, depressed, or disappointed, then the music style or background music of the first music is adjusted to a calm or soothing bias.

[0186] For example, the first scene description text is "The vehicle is driving on the coastal highway, the sun shines on the beach, birds fly in the air and chirp, and the waves crash against the shore." The first emotion description text is "The user's face is glaring angrily, their fists are clenched, and their voice is hoarse." Based on this first scene description text, the first emotion description text, and the artificial intelligence content generation model, the processing device determines that the first music's style is "rock music," its musical elements are "electronic keyboard, bass, electric guitar, and wave sounds," its accompaniment is "dynamic melody," its rhythm is strong and powerful, and its lyrics include one or more of the following: "seaside, highway, sunshine, beach, birds, flying, birdsong, waves." Thus, this application considers the user's emotions when generating the first music, making the generated music more in line with the user's emotional needs, providing personalized listening content, and improving cabin performance and the user's riding experience.

[0187] As another example, the processing device can also incorporate the user's preference information during the generation of the first piece of music. This preference information indicates the user's preferred music style, such as preferred instrument sounds, preferred background music, and preferred timbre. It should be noted that if the user provides explicit instructions and preferences as input, the generation of the first piece of music is biased towards the user's input; otherwise, the generation of the first piece of music relies on a subsequent AI content generation model. For example, the processing device obtains the first piece of music based on the user's preference information, the first scene description text, and the AI ​​content generation model. For example, the first scene description text is "The vehicle is driving on a coastal road, the sun shines on the beach, birds fly in the air and chirp, and the waves crash against the shore." Based on this first scene description text and an AI content generation model, the processing device determines the music style of the first piece of music to be "pop music," the musical elements to be "guitar sounds and wave sounds," the accompaniment to be "upbeat melody," the rhythm to be brisk, and the lyrics to include one or more of the following: "seaside, road, sunshine, beach, birds, flying, bird chirping, wave sounds." However, if the user's preference information indicates that the preferred instrument is the guzheng (Chinese zither) and the preferred music style is rock music, then the processing device will ultimately determine the music style of the first piece of music to be "rock music," and the musical elements to be "guzheng sounds and wave sounds." Thus, the first piece of music generated by this application better matches the user's preferences, providing personalized listening content for the in-vehicle user and improving cabin performance and user experience.

[0188] Optionally, the user's preference information can be user-inputted preference information. For example, the processing device receives an input first instruction, which indicates the user's preference information. For example, the user can manually input preference information on the in-vehicle system, thereby allowing the processing device to obtain the user's preference information. For instance, the user can manually input preference information on the in-vehicle system's user interface. As another example, the user can input preference information through voice interaction with the in-vehicle system, thereby allowing the processing device to obtain the user's preference information. For instance, the user can wake up the in-vehicle system's voice assistant and input preference information through voice interaction with the voice assistant.

[0189] Optionally, the user's preference information can be determined by the processing device based on the interaction log between the user and the vehicle. Optionally, the interaction log may include at least one of the user's driving information and the vehicle's historical music playback records. For example, the processing device can obtain the user's driving information, determine the user's driving style, and determine the user's preference information based on the user's driving style. For example, if the user's driving style is aggressive, the processing device can determine that the user's preferred music style might be energetic music, and the user's preferred music tempo might be fast-paced. As another example, the processing device can obtain the vehicle's historical music playback records and determine the user's preference information based on these records. For example, the processing device can search for music with high playback volume in the vehicle's historical music playback records and analyze the music style, tempo, etc., of the music with high playback volume to determine the user's preference information.

[0190] As another example, the processing device may also combine the first audio during the generation of the first music. For instance, the processing device obtains the first music based on the first audio, the first scene description text, and the artificial intelligence content generation model. For example, the first audio may include sounds such as rain, wind, waves, birdsong, and human laughter. Optionally, the first audio may be included in the background music or accompaniment of the first music.

[0191] Optionally, the processing device can perform noise reduction and extract audio segments from the sounds of the external environment and / or the user's voice inside the vehicle to obtain a first audio. Optionally, the sounds of the external environment and the user's voice inside the vehicle can be acquired by a collection device, such as the first sound data in the first input information. In this way, the generated first music is more in line with the scene in which the vehicle is located, which can enhance the user's immersion and experience.

[0192] As another example, the processing device can also incorporate location data during the generation of the first music. For instance, the first input information also includes first location data, and the processing device obtains the first music based on the first location data, the first scene description text, and the artificial intelligence content generation model. Optionally, the musical elements of the first music are related to the musical elements of the region indicated by the first location data. For example, when the first location data indicates that the vehicle is located in the Gobi Desert region of Northwest China, since the musical elements of the Gobi Desert region include instruments such as the morin khuur (horsehead fiddle), drums, and flutes, the musical elements of the first music can also include instruments such as the morin khuur, drums, and flutes. Similarly, when the first location data indicates that the vehicle is located in the Jiangnan water towns region of China, since the musical elements of the Jiangnan water towns region include instruments such as the guqin (zither) and xiao (vertical flute), the musical elements of the first music can also include instruments such as the guqin and xiao. Furthermore, when the first location data indicates that the vehicle is located in the Inner Mongolia grassland region of China, since the musical elements of the Inner Mongolia grassland region include vocal music such as throat singing, the musical elements of the first music can also include vocal music such as throat singing.

[0193] Since the vehicle may be in motion, and the scene it is in may be constantly changing, the processing device needs to acquire input information again. That is, after step S403, the processing device can also acquire second input information. Based on the second input information, a second scene description text is obtained. Based on the second scene description text and the artificial intelligence content generation model, a second piece of music is obtained. The second input information includes at least one of the following: second visual data and second vehicle speed data and second vehicle acceleration data. The second scene description text is used to indicate the second scene in which the vehicle is located. The second piece of music includes instrumental music or a song. The first scene and the second scene are different, and the first piece of music are different from the second piece of music.

[0194] Optionally, the difference between the first and second music pieces may include at least one of the following: the musical style of the first music piece is different from that of the second music piece; the musical elements of the first music piece are different from those of the second music piece; the accompaniment of the first music piece is different from that of the second music piece; the rhythm of the first music piece is different from that of the second music piece; and the lyrics of the first music piece are different from those of the second music piece.

[0195] For example, as shown in Figure 9, when the vehicle is traveling in a tunnel, the first scene description text obtained by the processing device is "The vehicle is traveling in a tunnel; the light is dim and the noise is loud." The music style of the first piece of music generated by the processing device is "pop music," and the musical elements of the first piece of music are "guitar and drum sounds," "dynamic melody," and "upbeat rhythm," among one or more of these. When the vehicle exits the tunnel, as shown in Figure 10, when the vehicle is traveling on a forest road, the second scene description text obtained by the processing device is "The vehicle is traveling on a forest road; trees line both sides of the road, and birds are flying among the trees, making bird calls." The music style of the second piece of music generated by the processing device is "light music," and the musical elements of the second piece of music are "piano sounds and bird calls," "soothing melody," and "upbeat rhythm," among one or more of these. In other words, the music generated by the processing device changes as the scene in which the vehicle is traveling changes. In this way, the music generated by this application can change dynamically with the changes in the external scene of the vehicle, that is, the sound moves with the scenery, which can provide personalized listening content for users in the car, giving users a highly immersive experience, satisfying people's needs for the intelligence and entertainment of the vehicle, and improving the performance of the cabin and the user's riding experience.

[0196] Since the first scene description text is obtained by the processing device based on various data in the first input information, and the second scene description text is obtained by the processing device based on various data in the second input information, when a certain data in the second input information changes compared to a certain data in the first input information, the obtained second scene description text may be different from the first scene description text, which in turn may cause the second music generated by the processing device to be different from the first music.

[0197] There are several possible scenarios where a certain data in the second input information differs from a certain data in the first input information, resulting in the second music generated by the processing device being different from the first music. Several possible scenarios are described below.

[0198] In scenario one, the first input information includes first visual data and the vehicle's first speed data, and the second input information includes second visual data and the vehicle's second speed data. The processing device extracts the same motion features from the first and second speed data, but extracts different visual features from the first and second visual data. For example, the processing device obtains the first visual features of the vehicle's exterior as "tunnel, dim lighting" based on the first visual data. The processing device obtains the second visual features of the vehicle's exterior as "forest, road, trees, birds, flying" based on the second visual data.

[0199] As can be seen, the second visual feature is different from the first visual feature, therefore the second music generated by the processing device based on the second input information is different from the first music generated based on the first input information. The process by which the processing device generates the second music based on the second input information, and the process by which the processing device generates the first music based on the first input information, are described in the aforementioned relevant descriptions and will not be repeated here.

[0200] For example, the musical styles of the first and second songs differ; for instance, the first song is "pop music," while the second song is "light music." As another example, the musical elements of the first and second songs differ; for example, the first song uses "guitar and drum sounds," while the second song uses "piano and birdsong." Furthermore, the accompaniment of the first and second songs differs; for example, the first song has a "dynamic melody," while the second song has a "soothing melody." And, as yet another example, the lyrics of the first and second songs differ; the first song's lyrics include "tunnel, dim," while the second song's lyrics include "forest, road, trees, birds, flying."

[0201] In scenario two, the first input information includes first visual data and the vehicle's first speed data, and the second input information includes second visual data and the vehicle's second speed data. The processing device extracts the same visual features from the first and second visual data, but extracts different motion features from the first and second speed data. For example, the processing device obtains the vehicle's third motion feature based on the first speed data as "the vehicle's speed is 100 km / h, and the vehicle is accelerating." The processing device obtains the vehicle's fourth motion feature based on the second speed data as "the vehicle's speed is 50 km / h, and the vehicle is moving at a constant speed."

[0202] It can be seen that the third motion feature is different from the fourth motion feature, so the second music generated by the processing device based on the second input information is different from the first music generated based on the first input information.

[0203] For example, the musical styles of the first and second pieces of music differ; for instance, the first piece is "pop music," while the second piece is "light music." As another example, the musical elements of the first and second pieces differ; for example, the first piece uses "guitar and drum sounds," while the second piece uses "piano and flute sounds." Furthermore, the accompaniment of the first and second pieces differs; for example, the first piece has a "cheerful melody," while the second piece has a "soothing melody." And yet another example, the rhythms of the first and second pieces differ; the first piece has a fast rhythm, while the second piece has a slow rhythm.

[0204] In scenario three, the first input information includes first visual data and the vehicle's first acceleration data, the second input information includes second visual data and the vehicle's second acceleration data, and the visual features extracted by the processing device based on the first visual data and the second visual data are the same, while the motion features extracted by the processing device based on the first acceleration data and the second acceleration data are different.

[0205] The vehicle's first acceleration data includes the vehicle's first vertical acceleration data and the vehicle's first horizontal acceleration data, while the vehicle's second acceleration data includes the vehicle's second vertical acceleration data and the vehicle's second horizontal acceleration data. For example, the processing device obtains the vehicle's fifth motion characteristic based on the first acceleration data as "the vehicle's vertical acceleration is 0.25 * 9.8 m / s²". 2 The vehicle was experiencing severe shaking, with a horizontal acceleration of 6 m / s². 2 The vehicle is in a state of rapid acceleration. Based on the second acceleration data, the processing device determines the vehicle's sixth motion characteristic as: "The vehicle's vertical acceleration is 0.1 * 9.8 m / s²." 2 The vehicle is in a stable state, and its horizontal acceleration is -1 m / s². 2 The vehicle is in a state of slow deceleration.

[0206] It can be seen that the fifth motion feature is different from the sixth motion feature, so the second music generated by the processing device based on the second input information is different from the first music generated based on the first input information.

[0207] For example, the musical styles of the first and second pieces of music differ; for instance, the first piece is "rock music," while the second piece is "light music." As another example, the musical elements of the first and second pieces differ; for example, the first piece uses "keyboard, bass, and electric guitar" sounds, while the second piece uses "piano and flute" sounds. As yet another example, the accompaniment of the first and second pieces differs; for instance, the first piece has a "dynamic melody," while the second piece has a "soothing melody." As yet another example, the rhythms of the first and second pieces differ; the first piece has a fast rhythm, while the second piece has a slow rhythm.

[0208] Scenario 4: The first input information includes at least one of the following: first visual data and vehicle's first speed data and vehicle's first acceleration data, and also includes vehicle's first ambient temperature data. The second input information includes at least one of the following: first visual data and vehicle's second speed data and vehicle's second acceleration data, and also includes vehicle's second ambient temperature data. Furthermore, the visual features extracted by the processing device based on the first visual data and the second visual data are the same; the motion features extracted by the processing device based on the first speed data and the second speed data are the same; the motion features extracted by the processing device based on the first acceleration data and the second acceleration data are the same; however, the environmental features extracted by the processing device based on the first ambient temperature data and the second ambient temperature data are different. For example, the third environmental feature obtained by the processing device based on the first ambient temperature data is "ambient temperature 30℃", and the fourth environmental feature obtained by the processing device based on the second ambient temperature data is "ambient temperature 10℃".

[0209] It can be seen that the third environmental feature is different from the fourth environmental feature, so the second music generated by the processing device based on the second input information is different from the first music generated based on the first input information.

[0210] For example, the musical styles of the first and second pieces of music differ; for instance, the first piece is "pop music," while the second piece is "light music." As another example, the musical elements of the first and second pieces differ; for example, the first piece uses "guitar and drum sounds," while the second piece uses "piano and flute sounds." Furthermore, the accompaniment of the first and second pieces differs; for example, the first piece has a "cheerful melody," while the second piece has a "soothing melody." And yet another example, the rhythms of the first and second pieces differ; the first piece has a fast rhythm, while the second piece has a slow rhythm.

[0211] Case 5: The first input information includes at least one of the following: first visual data and vehicle's first speed data and vehicle's first acceleration data, and also includes vehicle's first sound data. The second input information includes at least one of the following: first visual data and vehicle's second speed data and vehicle's second acceleration data, and also includes vehicle's second sound data. Furthermore, the visual features extracted by the processing device based on the first visual data and the second visual data are the same; the motion features extracted by the processing device based on the first speed data and the second speed data are the same; the motion features extracted by the processing device based on the first acceleration data and the second acceleration data are the same; however, the sound features extracted by the processing device based on the first sound data and the second sound data are different. For example, the first sound feature obtained by the processing device based on the first sound data is "wave sound and bird song," and the second sound feature obtained by the processing device based on the second sound data is "rain sound and wind sound."

[0212] It can be seen that the first sound feature is different from the second sound feature, so the second music generated by the processing device based on the second input information is different from the first music generated based on the first input information.

[0213] For example, the musical elements of the first piece of music are different from those of the second piece of music. For instance, the musical elements of the first piece of music are "the sound of waves and birdsong", while the musical elements of the second piece of music are "the sound of rain and wind".

[0214] The above example illustrates a situation where a certain data point in the second input information differs from a certain data point in the first input information, causing the processing device to obtain different first and second scene description texts, which in turn leads to the processing device generating different second and first music. Of course, in actual use, many more possibilities exist. For example, some data points in the second input information may differ from some data points in the first input information; that is, the above-mentioned situations can be combined. These will not be illustrated individually here.

[0215] Since the vehicle may be in motion, the emotions of the users inside the vehicle may change constantly. Therefore, the processing device needs to acquire new input information. That is, after step S403, the processing device can also acquire third input information. Based on the third input information, a second emotion description text is obtained. Based on the first scene description text, the second emotion description text, and the artificial intelligence content generation model, a third piece of music is obtained. The third input information includes at least one of a third image of the user inside the vehicle and a third sound of the user inside the vehicle. The second emotion description text is used to indicate the second emotion of the user inside the vehicle. The third piece of music includes instrumental music or a song, and the first emotion is different from the second emotion, and the first piece of music is different from the third piece of music.

[0216] Optionally, the difference between the first and third music includes at least one of the following: the musical style of the first music is different from that of the third music; the musical elements of the first music are different from those of the third music; the accompaniment of the first music is different from that of the third music; the rhythm of the first music is different from that of the third music; and the lyrics of the first music are different from those of the third music.

[0217] For example, the first emotion description text is "the user's face is beaming, their posture is laughing heartily, and they are emitting loud laughter." Based on the first scene description text, the first emotion description text, and the artificial intelligence content generation model, the processing device determines the music style of the first piece of music as "pop music," the music elements as "guitar sounds and ocean waves," the accompaniment as "upbeat melody," the rhythm as a brisk beat, and the lyrics as "seaside, highway, sunshine, beach, birds, flying, birdsong, ocean waves," etc. The second emotion description text is "the user's face is glaring angrily, their posture is clenched fists, and their voice is hoarse." Based on the first scene description text, the first emotion description text, and the artificial intelligence content generation model, the processing device determines the music style of the first piece of music as "rock music," the music elements as "electronic keyboard sounds, bass sounds, electric guitar sounds, and ocean waves," the accompaniment as "dynamic melody," the rhythm as a strong and powerful beat, and the lyrics as "seaside, highway, sunshine, beach, birds, flying, birdsong, ocean waves," etc. In this way, the music generated by this application can dynamically change according to the emotional changes of the users inside the vehicle, providing personalized listening content for the users inside the vehicle, giving them a highly immersive experience, satisfying people's needs for the intelligence and entertainment of the vehicle, and improving the performance of the cabin and the user's riding experience.

[0218] Since the first emotion description text is obtained by the processing device based on various data in the first input information, and the second emotion description text is obtained by the processing device based on various data in the third input information, when a certain data in the third input information changes compared to a certain data in the first input information, the resulting second emotion description text may be different from the first emotion description text, which in turn may cause the third music generated by the processing device to be different from the first music.

[0219] There are several possible scenarios where a certain data in the third input information differs from a certain data in the first input information, resulting in the third music generated by the processing device being different from the first music. Two possible examples are described below.

[0220] For example, the first input information includes at least one of first visual data and first speed data and first acceleration data of the vehicle, and the first visual data includes a second image of the user inside the vehicle. The third input information includes at least one of third visual data and third speed data and third acceleration data of the vehicle, and the third visual data includes a third image of the user inside the vehicle. Furthermore, the motion features extracted by the processing device based on the first speed data and the third speed data are the same, the motion features extracted by the processing device based on the first acceleration data and the third acceleration data are the same, but the visual features of the vehicle interior extracted by the processing device based on the first visual data and the third visual data are different. For example, the visual features of the vehicle interior obtained by the processing device based on the first visual data (referred to as the first emotional feature for easy distinction) are "a smiling face and a posture of laughing heartily", and the visual features of the vehicle interior obtained by the processing device based on the third visual data (referred to as the second emotional feature for easy distinction) are "a dull face and a posture of weeping".

[0221] It can be seen that the first emotional feature is different from the second emotional feature, so the third music generated by the processing device based on the third input information is different from the first music generated based on the first input information.

[0222] For example, the musical elements of the first piece are different from those of the third piece. For instance, the musical element of the first piece is "laughter," while the musical element of the third piece is "rain." As another example, the accompaniment of the first piece is a "cheerful melody," while the accompaniment of the third piece is a "melancholy melody." Yet another example, the rhythm of the first piece is a brisk rhythm, while the rhythm of the third piece is a slow rhythm.

[0223] For example, the first input information includes at least one of first visual data and first vehicle speed data and first vehicle acceleration data, and the first visual data includes a second voice of the user inside the vehicle. The third input information includes at least one of third visual data and third vehicle speed data and third vehicle acceleration data, and the third visual data includes a third voice of the user inside the vehicle. Furthermore, the motion features extracted by the processing device based on the first speed data and the third speed data are the same, the motion features extracted by the processing device based on the first acceleration data and the third acceleration data are the same, but the sound features extracted by the processing device based on the first voice and the third voice are different. For example, the sound feature inside the vehicle obtained by the processing device based on the first voice (for easy distinction, referred to as the third emotional feature) is "loud laughter", and the sound feature inside the vehicle obtained by the processing device based on the third voice (for easy distinction, referred to as the fourth emotional feature) is "wailing".

[0224] It can be seen that the third emotional feature is different from the fourth emotional feature, so the third music generated by the processing device based on the third input information is different from the first music generated based on the first input information.

[0225] For example, the musical elements of the first piece are different from those of the third piece. For instance, the musical element of the first piece is "laughter," while the musical element of the third piece is "sobbing." As another example, the accompaniment of the first piece is a "cheerful melody," while the accompaniment of the third piece is a "sad melody." As yet another example, the rhythm of the first piece is a brisk rhythm, while the rhythm of the third piece is a slow rhythm.

[0226] The above example illustrates a situation where a certain data point in the third input information differs from a certain data point in the first input information, causing the processing device to obtain different first and second emotion description texts, which in turn leads to the processing device generating different third music compared to the first music. Of course, in actual use, many more possibilities exist. For example, some data points in the third input information may differ from some data points in the first input information. In other words, the above examples can be combined with each other, which will not be elaborated upon here.

[0227] Optionally, in actual use, the vehicle's environment may change constantly, as may the emotions of the users inside the vehicle. The processing device can also combine the changed vehicle environment and the changed emotions of the users inside the vehicle to generate new music. For example, the processing device can generate a fourth piece of music based on the second scene description text, the second emotion description text, and the artificial intelligence content generation model. The fourth piece of music differs from the first piece. Similar to the previous description, it will not be explained in detail here. Thus, the music generated by this application can dynamically change with the vehicle's environment and the emotions of the users inside the vehicle, making the generated music more consistent with the vehicle's environment and the users' emotional needs. This provides personalized listening content for users inside the vehicle, giving them a highly immersive experience, satisfying their needs for vehicle intelligence and entertainment, and improving cabin performance and the user's riding experience.

[0228] Optionally, the processing device can also determine the user's timbre based on the user's voice inside the vehicle, and generate the first piece of music based on the user's timbre, the first scene description text, and the artificial intelligence content generation model. In other words, the timbre of the singer in the first piece of music can be the user's timbre, making the generated first piece of music sound more like it was sung by the user themselves. This can improve the playability of the generated first piece of music, provide personalized listening content for users inside the vehicle, give users a highly immersive experience, and enhance the user's riding experience.

[0229] Furthermore, after obtaining the first music (i.e., after step S403), the processing device can also output the obtained first music and / or content related to the first music to the user via an interactive device. The interactive device is described in the foregoing related description.

[0230] As an example of visual output, an interactive device that outputs content related to the first music may include a screen that can display an interface capable of outputting the first music and / or content related to the first music in the form of text, symbols, or graphics.

[0231] For example, after receiving the first music, the processing device can generate a second interface and control the vehicle's interactive device to display the second interface. The second interface includes content related to the first music and a first control, which adjusts at least one of the following: the music style of the first music, the music elements of the first music, the background music of the first music, the music rhythm of the first music, and the lyrics of the first music.

[0232] As shown in Figure 11, the second interface includes content related to the first music and a first control. The content related to the first music is "fragment 1 of the first music, fragment 2 of the first music, the music style is pop music, the music elements are guitar sounds and ocean waves, the accompaniment is a light melody, and the rhythm is a brisk rhythm." The interactive device can adjust the content related to the first music through the first control based on the user's voice input (e.g., "adjust the music style of the first music to rock music"). That is, the content related to the first music is updated to "fragment 1 of the first music, fragment 2 of the first music, the music style is rock music, the music elements are guitar sounds and ocean waves, the accompaniment is a light melody, and the rhythm is a brisk rhythm." In this way, the user can adjust the first music obtained by the processing device through the interactive device, making the adjusted first music more in line with the user's preferences and intentions. This can enhance the user's immersion, increase the playability of the generated music, and improve the user's participation and user experience.

[0233] Optionally, users can use the first control to adjust content related to the first music through input methods such as voice input or touch screen input.

[0234] For example, the first piece of music may include multiple music segments. Users can adjust a single music segment or a single lyric in the first piece of music through the first control. For example, users can select a single music segment or a single lyric in the first piece of music that they want to adjust through touch screen, or input the adjusted lyrics through voice input or touch screen input, etc.

[0235] Furthermore, after obtaining the first music (i.e., after step S403), the processing device can also generate a first video or a fourth image based on the first scene description text and the artificial intelligence content generation model, and output the obtained first video or fourth image to the user through an interactive device. The interactive device is described in the foregoing related description.

[0236] As an example of visual output, an interactive device that outputs a first video or a fourth image may include a screen that can display an interface capable of outputting the first video or fourth image in the form of text, symbols, or graphics.

[0237] For example, after obtaining the first music, the processing device can obtain a first video or a fourth image based on the first scene description text and the artificial intelligence content generation model, generate a third interface, and control the vehicle's interactive device to display the third interface. The third interface includes the first video or the fourth image. Optionally, the first video or the fourth image can be an animated video, cartoon image, simplified video, simplified image, etc., virtually extracted by the processing device based on the first scene description text and the artificial intelligence content generation model.

[0238] Taking the image generated by the processing device as a simplified image as an example, as shown in Figure 12, the third interface includes a fifth image. The fifth image indicates that the first scene in which the vehicle is located is "the vehicle is driving on the road by the sea, the sun shines on the beach, birds fly in the air and make bird calls, and the waves are crashing on the beach and making the sound of waves".

[0239] Furthermore, the processing device can combine the second interface with the third interface. For example, the processing device can use the third interface as a background image or background video and display the second interface on the background image or background video.

[0240] Taking the processing device combining the second and third interfaces to generate a fourth interface as an example, as shown in Figure 13, the fourth interface includes a fifth image, which serves as the background image of the fourth interface. The fourth interface also includes content related to the first music and a first control. The content related to the first music consists of "fragment 1 of the first music, fragment 2 of the first music, the music style is pop music, the music elements are guitar sounds and ocean waves, the accompaniment is a light melody, and the rhythm is a brisk rhythm." The first control is described in the aforementioned related description. Thus, this application allows users to view images or videos related to the first music while accessing or listening to it, enhancing user immersion, increasing the playability of music generation, and improving user participation and user experience.

[0241] In one possible implementation, this application can also be applied to other fields. For example, the processing device acquires input information such as video footage captured by a smartphone. By analyzing and understanding the images in the video and combining them with user input commands, the processing device generates background music that fits the video content using an artificial intelligence content generation model. As another example, in a smart home scenario, smart devices can acquire videos or images of the user and generate background music by recognizing the user's emotions.

[0242] It should be noted that the collection of input information in this application, such as the collection of all audio and image data, requires the consent of the user, and the storage, processing, or transmission of all data must comply with local laws and regulations.

[0243] In the embodiment shown in Figure 4, the processing device can acquire input information, obtain scene description text based on the input information, and obtain first music based on the scene description text and the artificial intelligence content generation model. The scene description text indicates the scene in which the vehicle is located. This application can generate music that matches the scene in which the vehicle is located based on input information and the artificial intelligence content generation model, which can improve the richness, creativity, and uniqueness of the generated music, bringing a more accurate and convenient experience to users. Moreover, music generated through the AIGC model avoids copyright issues and can reduce the cost of music generation.

[0244] The methods of the embodiments of this application have been described in detail above. Below, some apparatuses for implementing the foregoing methods are described. It should be understood that the division of units in the apparatuses provided in the embodiments of this application is only a logical functional division; in actual implementation, they can be fully or partially integrated onto a single physical entity, or they can be physically separated.

[0245] Furthermore, the units in the device can be implemented in the form of processor calling software. For example, the device includes a processor connected to a memory containing instructions. The processor calls the instructions stored in the memory to implement any of the above methods or to implement the functions of each unit of the device. The processor is, for example, a general-purpose processor, such as a central processing unit (CPU) or a microprocessor, and the memory is either internal or external to the device.

[0246] Alternatively, the units in the device can be implemented as hardware circuits. The functionality of some or all units can be achieved through the design of these hardware circuits, which can be understood as one or more processors. For example, in one implementation, the hardware circuit is an application-specific integrated circuit (ASIC). The functionality of some or all of the above units is achieved through the design of the logical relationships between the components within the circuit. In another implementation, the hardware circuit can be implemented using a programmable logic device (PLD). Taking a field-programmable gate array (FPGA) as an example, it can include a large number of logic gates. The connection relationships between the logic gates are configured through a configuration file, thereby achieving the functionality of some or all of the above units. All units of the above device can be implemented entirely through processor-invoked software, entirely through hardware circuits, or partially through processor-invoked software with the remaining parts implemented through hardware circuits.

[0247] In this application embodiment, a processor is a circuit with signal processing capabilities. In one implementation, the processor can be a circuit with instruction read and execute capabilities, such as a central processing unit (CPU) or a digital signal processor (DSP). In another implementation, the processor can implement certain functions through the logical relationships of hardware circuits. These logical relationships are fixed or reconfigurable. For example, the processor is a hardware circuit implemented using an application-specific integrated circuit (ASIC) or a programmable logic device (PLD), such as an FPGA. In a reconfigurable hardware circuit, the process of the processor loading a configuration document and configuring the hardware circuit can be understood as the process of the processor loading instructions to implement the functions of some or all of the above units. Therefore, each unit in the device can be one or more processors (or processing circuits) configured to implement the above methods, such as a CPU, GPU, microprocessor, DSP, ASIC, FPGA, or a combination of at least two of these processor types.

[0248] Furthermore, the units in the above devices can be integrated in whole or in part, or they can be implemented independently. In one implementation, these units are integrated together as a system-on-a-chip (SOC). The SOC may include at least one processor for implementing any of the above methods or for implementing the functions of the units in the device. The at least one processor may be of different types, such as a CPU and an FPGA.

[0249] Several possible devices are listed below.

[0250] Please refer to Figure 14, which is a schematic diagram of a processing device provided in an embodiment of this application, namely processing device 140. Optionally, the processing device 140 can be a standalone device, such as the processing device 102 shown in Figure 1. Alternatively, the processing device 140 can also be a component in a standalone device (such as a node), such as a chip or integrated circuit. The processing device 140 is used to implement the music generation method shown in Figure 4 above.

[0251] As shown in Figure 14, the processing device 140 includes an acquisition unit 1401 and a processing unit 1402. The acquisition unit 1401 is used to perform one or more operations such as acquisition, reception, listening, transmission, and sending, for example, to acquire first input information. The first input information includes first visual data and at least one of the following: first speed data of the vehicle and first acceleration data of the vehicle. It further includes other operations for implementing the music generation method.

[0252] The processing unit 1402 is used to perform one or more operations such as processing, calculation, determination, generation, and updating. For example, it is used to obtain a first scene description text based on the first input information, and to obtain first music based on the first scene description text and an artificial intelligence content generation model. The first scene description text indicates the first scene in which the vehicle is located, and the first music includes instrumental music or a song. It further includes other operations for implementing the music generation method.

[0253] For related descriptions, please refer to the description of the embodiment shown in Figure 4, which will not be described in detail here.

[0254] Please refer to Figure 15, which is a schematic diagram of the structure of a computing device provided in an embodiment of this application. A computing device is a device with processing capabilities. The device here can be a physical device, such as a server (e.g., a rack server) or a host, or it can be a virtual device, such as a virtual machine or a container.

[0255] As shown in Figure 15, the computing device 150 includes a processor 1501, a memory 1502, and one or more programs, and may include a communication interface 1503. It should be understood that this application does not limit the number of processors and memories in the computing device 150.

[0256] Processor 1501 is a module for performing calculations and may include a central processing unit (CPU), a graphics processing unit (GPU), a microprocessor (MP), a digital signal processor (DSP), a micro controller unit (MCU), or one or more integrated circuits for controlling the execution of programs in the above schemes.

[0257] Memory 1502 provides storage space, in which application data, user data, operating system, and computer programs can be optionally stored. Memory 1502 may include read-only memory (ROM) or other types of static storage devices capable of storing static information and instructions, random access memory (RAM) or other types of dynamic storage devices capable of storing information and instructions, or electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but is not limited thereto.

[0258] The memory 1502 can exist independently and be connected to the processor 1501 via a bus. Alternatively, the memory 1502 can be integrated with the processor 1501.

[0259] The communication interface 1503 is used to provide information input or output to the at least one processor. And / or, the communication interface 1503 can be used to receive data transmitted externally and / or transmit data externally. The communication interface 1503 can be a wired link interface, including an Ethernet cable, or a wireless link interface (Bluetooth, general wireless transmission, and other wireless communication technologies, etc.). Optionally, the communication interface 1503 may also include a transmitter (such as a radio frequency transmitter, antenna, etc.) or a receiver coupled to the interface.

[0260] In this embodiment, one or more programs are stored in the memory 1502 in the form of program code and configured to be executed by the processor 1501. The programs include instructions for implementing the steps in the music generation method shown in FIG4. That is, the memory 1502 stores executable instructions, and the processor 1501 executes the executable instructions to implement the steps in the music generation method shown in FIG4. In other words, the memory 1502 stores instructions for executing the music generation method shown in FIG4.

[0261] This application embodiment also provides a vehicle, which includes the aforementioned processing device 140 or computing device 150, and is used to implement the aforementioned music generation method, such as the music generation method shown in FIG4.

[0262] This application also provides a computer program product containing instructions. The computer program product may be a software or program product containing instructions, capable of running on a computing device or stored on any usable medium. The computer program instructions are used to implement the aforementioned music generation method, such as the music generation method shown in Figure 4.

[0263] This application also provides a computer-readable storage medium. This computer-readable storage medium is used to store a computer program, the computer program including instructions for implementing the aforementioned music generation method, such as the music generation method shown in FIG4.

[0264] The computer-readable storage medium can be any available medium that can be stored by an information interaction device and / or computing device, or a data storage device such as a data center containing one or more available media. The available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media, or semiconductor media (e.g., solid-state drives).

[0265] In this application, the terms "exemplary" or "for example" are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" or "for example" in this application should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of terms such as "exemplary" or "for example" is intended to present the relevant concepts in a specific manner.

[0266] In this application, "at least one" in the embodiments refers to one or more items, and "more than one" refers to two or more items. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of single or multiple items. For example, at least one of a, b, or c can represent: a, b, c, (a and b), (a and c), (b and c), or (a and b and c), where a, b, and c can be single or multiple. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, and B alone, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship.

[0267] Furthermore, unless otherwise stated, the use of ordinal numbers such as "first" and "second" in the embodiments of this application is for distinguishing multiple objects and is not for limiting the order, sequence, priority or importance of multiple objects.

[0268] Those skilled in the art will understand that all or part of the steps of the above embodiments can be implemented by hardware or by a program instructing related hardware. The program can be stored in a computer-readable storage medium, such as a read-only memory, a disk, or an optical disk.

[0269] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the protection scope of the technical solutions of the embodiments of this application.

Claims

1. A method for generating music, characterized in that, The method includes: Acquire first input information, the first input information including first visual data and at least one of the following: first speed data of the vehicle and first acceleration data of the vehicle; Based on the first input information, a first scene description text is obtained, which is used to indicate the first scene in which the vehicle is located; Based on the first scene description text and the artificial intelligence content generation model, the first music is obtained, which includes instrumental music or songs.

2. The method according to claim 1, characterized in that, The song includes music and lyrics.

3. The method according to claim 1 or 2, characterized in that, After obtaining the first piece of music based on the first scene description text and the artificial intelligence content generation model, the method further includes: Acquire second input information, the second input information including second visual data and at least one of the following: second speed data of the vehicle and second acceleration data of the vehicle; Based on the second input information, a second scene description text is obtained, which is used to indicate the second scene in which the vehicle is located; Based on the second scene description text and the artificial intelligence content generation model, a second piece of music is obtained, which includes instrumental music or songs. The first scene is different from the second scene, and the first music is different from the second music.

4. The method according to claim 3, characterized in that, The first music differs from the second music in at least one of the following ways: The musical styles of the first music and the second music are different; the musical elements of the first music and the second music are different; the accompaniment of the first music and the second music are different; the rhythm of the first music and the second music are different; and the lyrics of the first music and the second music are different.

5. The method according to any one of claims 1-4, characterized in that, The first visual data includes a first image of the environment outside the vehicle and / or a second image of the user inside the vehicle.

6. The method according to any one of claims 1-5, characterized in that, The first input information also includes the first ambient temperature data of the vehicle.

7. The method according to any one of claims 1-6, characterized in that, The first input information also includes first sound data, which includes first sounds of the environment outside the vehicle and / or second sounds of the user inside the vehicle.

8. The method according to any one of claims 1-7, characterized in that, The first input information also includes first location data, and the musical elements of the first music are related to the musical elements of the region indicated by the first location data.

9. The method according to any one of claims 1-8, characterized in that, The step of obtaining the first scene description text based on the first input information includes: Based on the first input information, one or more scene features are obtained; The first scene description text is obtained based on one or more scene features and the neural network.

10. The method according to claim 9, characterized in that, The step of obtaining the first scene description text based on one or more scene features and a neural network includes: The first scene description text is obtained based on the scene feature vector of the vehicle at the current moment, the scene feature vector of the vehicle at the previous moment, and the neural network. The scene feature vector of the vehicle at the current moment corresponds to the scene feature vector of the vehicle at the previous moment, and the scene feature vector is related to one or more scene features.

11. The method according to any one of claims 1-10, characterized in that, The time interval between the sampling time of the first input information and the time of the last acquisition of input information is related to the first speed data of the vehicle. or, The sampling duration of the first input information is related to the first speed data of the vehicle.

12. The method according to any one of claims 1-11, characterized in that, When the first input information includes at least one of a second image of a user inside the vehicle and a second voice of a user inside the vehicle, the method further includes: Based on the first input information, a first emotion description text is obtained, which is used to indicate the first emotion of the user inside the vehicle. The step of obtaining the first music based on the first scene description text and the artificial intelligence content generation model includes: The first piece of music is obtained based on the first scene description text, the first emotion description text, and the artificial intelligence content generation model.

13. The method according to claim 12, characterized in that, After obtaining the first piece of music based on the first scene description text, the first emotion description text, and the artificial intelligence content generation model, the method further includes: Obtain third input information, the third input information including at least one of a third image of a user inside the vehicle and a third voice of a user inside the vehicle; Based on the third input information, a second emotion description text is obtained, which is used to indicate the second emotion of the user inside the vehicle. Based on the first scene description text, the second emotion description text, and the artificial intelligence content generation model, a third piece of music is obtained, which includes instrumental music or songs. The first emotion is different from the second emotion, and the first music is different from the third music.

14. The method according to claim 12 or 13, characterized in that, The step of obtaining the first emotion description text based on the first input information includes: Based on the first input information, one or more emotion features are obtained; The first emotion description text is obtained based on one or more emotion features and the neural network.

15. The method according to claim 14, characterized in that, The step of obtaining the first emotion description text based on one or more emotion features and a neural network includes: The first emotion description text is obtained based on the vehicle's emotion feature vector at the current moment, the vehicle's emotion feature vector at the previous moment, and the neural network. The vehicle's emotion feature vector at the current moment corresponds to the vehicle's emotion feature vector at the previous moment, and the emotion feature vector is related to one or more emotion features.

16. The method according to any one of claims 1-15, characterized in that, The method further includes: Obtain the user's preference information for the vehicle, the preference information being used to indicate the user's music preference type; The step of obtaining the first music based on the first scene description text and the artificial intelligence content generation model includes: The first piece of music is obtained based on the preference information, the first scene description text, and the artificial intelligence content generation model.

17. The method according to claim 16, characterized in that, The process of obtaining the user's preference information for the vehicle includes: Receive a first instruction input, the first instruction being used to indicate the user's preference information; or, Based on the interaction log between the user and the vehicle, the user's preference information is determined. The interaction log includes at least one of the user's driving information and the vehicle's historical music playback records.

18. The method according to any one of claims 1-17, characterized in that, The method further includes: Get the first audio; The step of obtaining the first music based on the first scene description text and the artificial intelligence content generation model includes: The first piece of music is obtained based on the first audio, the first scene description text, and the artificial intelligence content generation model.

19. The method according to any one of claims 1-18, characterized in that, After obtaining the first scene description text based on the first input information, the method further includes: The system displays a first interface, which includes content related to the first scene description text and a first input area. The first input area is used to instruct the user of the vehicle to adjust the content related to the first scene description text.

20. The method according to any one of claims 1-19, characterized in that, After obtaining the first piece of music based on the first scene description text and the artificial intelligence content generation model, the method further includes: A second interface is displayed, the second interface including content related to the first music and a first control, the first control being used to adjust at least one of the following: The musical style of the first music, the musical elements of the first music, the background music of the first music, the musical rhythm of the first music, and the lyrics of the first music.

21. The method according to any one of claims 1-20, characterized in that, After obtaining the first piece of music based on the first scene description text and the artificial intelligence content generation model, the method further includes: Based on the first scene description text and the artificial intelligence content generation model, a first video or a fourth image is obtained; A third interface is displayed, which includes the first video or the fourth image.

22. A processing apparatus, characterized in that, The processing device includes: An acquisition unit is configured to acquire first input information, the first input information including first visual data and at least one of the following: first speed data of the vehicle and first acceleration data of the vehicle; Processing unit, used for: Based on the first input information, a first scene description text is obtained, which is used to indicate the first scene in which the vehicle is located; Based on the first scene description text and the artificial intelligence content generation model, the first music is obtained, which includes instrumental music or songs.

23. A computing device, characterized in that, The computing device includes a processor and a memory, the memory storing a program, and the processor executing the program to cause the computing device to implement the method as described in any one of claims 1-21.

24. A vehicle, characterized in that, The vehicle includes the processing device as described in claim 22 or the computing device as described in claim 23.

25. A computer-readable storage medium, characterized in that, The computer-readable storage medium is used to store a computer program, the computer program including instructions for performing the method as described in any one of claims 1-21.

26. A computer program product, characterized in that, The computer program product includes instructions that, when executed by a processor, cause the method as described in any one of claims 1-21 to be implemented.

Images